Socioeconomic disparities in cardiovascular (CV) disease are evidenced by a 2.5 times higher risk of cardiovascular mortality for low-income Americans, when compared to those in the highest income bracket. Despite well-documented awareness that social environments can influence CV disease risk, and the increasing attention given to disparities research, there is relatively little knowledge of the underlying social pathways contributing to socioeconomic disparities, and socioeconomic disparities in CV disease have only widened in the last three decades. Socioeconomic disparities exist for many CV risk behaviors, including smoking, lack of physical activity and insufficient fruit and vegetable intake, and these disparities contribute to socioeconomic gradients in CV disease. These behaviors are shaped by social environmental resource factors, including social support, loneliness and neighborhood social cohesion. These three social environmental resource measures are independently associated with CV morbidity, but their role in CV risk behaviors is not well understood. Also, interaction between social resources is likely to have a "multiplier effect", where the net effect is greater than the sum of the individual effects. However, few studies utilize an ecological framework to identify the magnitude of these effects simultaneously. Furthermore, these social resource measures are often correlated with socioeconomic position (SEP), indicating that the resources may represent a fundamental social inequality. The purpose of this study is to examine the relationships among SEP and three distinct, but related, measures of social resources - social support (both emotional and instrumental), loneliness and neighborhood social cohesion - with CV risk behavior outcomes (smoking, lack of physical activity, and insufficient fruit and vegetable intake). The aims of this study are: 1) To examine associations of social support, loneliness and neighborhood social cohesion with CV risk behaviors after controlling for confounders. 2) To explore the interactions of social support, loneliness and neighborhood social cohesion as they relate to CV risk behaviors. 3) To determine the contributions of social support, loneliness and neighborhood social cohesion to SEP gradients in CV risk behaviors. This study will use cross-sectional data obtained by the Multi-Ethnic Study of Atherosclerosis, a prospective cohort study of a diverse, population-based sample of 6814 adults without apparent CVD from six US sites. Multi-level statistical modeling techniques will be used to test the proposed inter-relationships of SEP and the three social resource measures, based on an ecological framework. PUBLIC HEALTH RELEVANCE: Results of this study will elucidate the pathways that contribute to socioeconomic CV risk behavior disparities, responding to a NINR research priority area, and this knowledge may inform future research into other health disparities. This knowledge also is important for the development of socially and contextually based behavioral interventions for clinicians and researchers, meeting a critical need in health behavior research. Moreover, these findings may be used to improve patient advocacy and shape policies to address health disparities at a societal level.